PROJECT SUMMARY Protein kinase inhibitors are potent anti-cancer therapeutics. However, many patients develop resistance against these inhibitors. For example, the Bcr-Abl kinase inhibitor imatinib decreases mortality for Chronic Myeloid Leukemia (CML) by 80%, but 22-41 % of patients acquire resistance to imatinib during treatment. The majority of relapsed patients harbor mutations in the Bcr-Abl kinase domain (KD), where more than 90 different mutations have been identified. Many of these patient-derived resistance mutations show no change in equilib- rium affinity for imatinib towards Abl kinase and are therefore expected to alter the imatinib binding process by other mechanisms. Mutations that affect the binding kinetics of imatinib by binding and dissociating rapidly may retain a high affinity for imatinib under equilibrium conditions. However, in the non-equilibrium environment of the cell or human body, mutations that increase the drug dissociation rate from its target can presumably con- fer resistance by reducing drug residence time. Not surprisingly, the concept of drug residence time has emerged as a superior predictor of cellular drug efficacy. Recent structure and dynamics experiments have shown that association and dissociation rates of drugs to their target kinases can be limited by the accessibility of the binding site, highlighting the importance of studying the process of drug binding to proteins. Additionally, our recent simulations of imatinib binding to Abl have revealed that Abl kinase accesses a transient conforma- tional state where it becomes partially unfolded. Presumably, mutations that destabilize Abl kinase, mimicking the partial unfolding that we observe in our simulations, would increase association/dissociation rates of imatinib and potentially confer drug resistance. Furthermore, chaperone machinery (hsp90/cdc37) in the cell is able to buffer structurally destabilizing mutations, including Bcr-Abl mutants, assisting cancer cells by decreas- ing the availability of the kinase to be degraded. We have identified patient-derived resistance mutations that show no change in equilibrium affinity for imatinib, but rather affect the stability of the protein. Taken together, I hypothesize that these clinical mutations in Abl kinase domain can cause imatinib resistance by increasing the dissociation rate and by tightening the interaction between Abl kinase and Hsp90. This hypothesis will be tested through the design of specific resistance mutations on Abl kinase and determination of the resulting changes in ligand binding rates and affinities determined by stopped-flow kinetics, surface plasmon resonance (SPR), fluorescence spectroscopy, NMR, and isothermal titration calorimetry (ITC) experiments. The results of this proposal are expected to provide insights into the mechanism of ?kinetic? resistance mutations. On a broader level, this study will show how ligands find their binding sites on proteins, leading to new strategies for improving drug specificity and efficacy by incorporating binding kinetics in the design of therapeutics. Addition- ally, we will gain a better understanding of how rates of conformational changes in protein kinases relate to their function and regulation.